Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    instances of specific words such as "date," "test," and "concluded."

    oai_token-act-pair · gpt-3.5-turbo

    instances of uncertainty or ambiguity in events or information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ain
    -0.78
     damn
    -0.67
    Ŀ
    -0.65
     darn
    -0.65
     NEVER
    -0.64
     mediocre
    -0.63
     ALWAYS
    -0.63
     crap
    -0.61
     ye
    -0.61
     respectable
    -0.61
    POSITIVE LOGITS
     nor
    1.32
     specifics
    0.97
     or
    0.96
     specific
    0.88
    nor
    0.87
     anymore
    0.85
    etz
    0.72
     particulars
    0.68
     Or
    0.68
    Or
    0.67
    Act Density 0.584%

    No Known Activations